Abstract: A method for recognizing a dialogue intention, an electronic device and a storage medium are provided. The solution includes: obtaining a sample sentence with an intention category to be predicted and a plurality of supporting sentences each labeled with an intention category; and inputting the sample sentence and the plurality of supporting sentences into a trained intention recognition model, in which the trained intention recognition model is configured to generate a sample sentence vector corresponding to the sample sentence and category vectors corresponding to the plurality of supporting sentences based on the sample sentence and the plurality of supporting sentences, calculate matching degrees between the sample sentence vector and the category vectors, and obtain a predicted intention category of the sample sentence based on the matching degrees.

Abstract: Methods, apparatuses, electronic devices, and mediums related to classification models are provided. The method may include: obtaining a set of positive samples and a set of negative samples, a sample in the set of positive samples including a pair of images from a first image set annotated as similar with each other, and a sample in the set of negative samples including a pair of images annotated as dissimilar with each other from the first image set; training a first sub-network based on the set of positive samples and the set of negative samples; constructing a classification model including a second sub-network, where initial weights of the second sub-network are weights of corresponding nodes of the trained first sub-network; and training the classification model using a second image set that includes at least one image that is common with at least one image in the first image set.

Abstract: Provided are methods and apparatuses for navigation guidance and establishing a three-dimensional real scene model, a device and a medium, which relate to the field of artificial intelligence and, in particular, intelligent transportation technologies.

Abstract: A method and a system for arranging business process, a computing device, and a non-transitory computer readable storage medium are provided. The method includes: receiving an application module and a business process rule that are transmitted by a user terminal, wherein the application module is configured to indicate a processing logic in a link in a business process, and wherein the business process rule is configured to indicate a processing rule of the application module; determining a target edge device corresponding to the application module; and transmitting the application module and the business process rule to the target edge device for the target edge device to execute the application module according to the business process rule.

Abstract: The present application discloses a method and apparatus for identifying an updated road, a device and a computer storage medium, and relates to the field of big data technologies. A specific implementation solution is as follows: comparing a road area extracted based on the latest satellite image with a road area extracted based on a historical satellite image, to obtain a candidate updated road; mapping the candidate updated road into road network data according to a coordinate position of the candidate updated road; acquiring a user trajectory set corresponding to the candidate updated road within a recent preset period; and identifying, based on a matching result between the user trajectory set and the road network data, whether the candidate updated road is an actual updated road. Updated roads can be more accurately identified through the method according to the present application.

Abstract: Provided are a method and apparatus for detecting a status of a lens, an electronic device and a medium. A specific implementation includes: determining current feature points included in a current image according to the current image captured at a current moment by a to-be-detected lens of a target driving device; determining historical feature points included in a historical image according to the historical image captured at a historical moment by the to-be-detected lens; and matching the current feature points with the historical feature points and determining a status detection result of the to-be-detected lens according to a matching result. Therefore, an effect of determining a status of the to-be-detected lens of the target driving device is achieved and the problem of a potential safety hazard of the target driving device due to an abnormal status of the to-be-detected lens can be avoided.

Abstract: Technical solutions relate to the field of artificial intelligence such as deep learning, computer vision and intelligent imaging. A method may includes during training of a one-stage object detecting model, acquiring values of a loss function corresponding to feature maps at different scales respectively in the case that classification loss calculation is required, and the loss function is a focal loss function; and determining a final value of the loss function according to the acquired values of the loss function, and training the one-stage object detecting model according to the final value of the loss function.

Abstract: A method of processing audio data, which relates to a field of speech synthesis technology. The method includes: decomposing original audio data to obtain voice audio data and background audio data; performing electroacoustic processing on the voice audio data to obtain electroacoustic voice data; and combining the electroacoustic voice data and the background audio data to obtain target audio data. An electronic device and a storage medium are further provided.

Abstract: A human-machine interaction method, includes: determining a target controlled object corresponding to a first gesture in a first area of a touch pad in response to detecting the first gesture in the first area; determining a target control mode corresponding to a second gesture in a second area of the touch pad in response to detecting the second gesture in the second area; and controlling the target controlled object based on the target control mode.

Abstract: A method of pushing a video, an electronic device and a storage medium are provided, which relate to a field of data processing technology, in particular to big data technology. The method includes: acquiring first high-quality videos; acquiring posterior data obtained when each first high-quality video is distributed for a first preset number of times as a label for the video; acquiring a first preset number of users earliest interacting with each first high-quality video; generating training samples using the videos as video-side features and the users as user-side features to train at least one ranking model; acquiring a first target user; acquiring one or more candidate videos corresponding to each opinion leader to form a first video set; and pushing, after ranking for the first video set using the at least one trained ranking model, a top second preset number of videos to the first target user.

Abstract: A method of protecting a model, which relates to a field of computer, a field of artificial intelligence, and may be applied to an AI model protection scenarios. The method includes: generating a WASM file for providing a runtime environment for a target model, the WASM file containing a corresponding model inference algorithm and security verification algorithm, wherein the security verification algorithm is configured to perform at least one security verification operation to protect the target model, the at least one security verification operation is selected from: a verification of a host environment; a verification of an integrity of the WASM file; a verification of an integrity of the model file generated corresponding to an original model file of the target model; a timeout verification of a specified inference process during a model inference process; or a timeout verification of an entire inference process during the model inference process.

Abstract: A file moving method, an electronic device, and a computer-readable storage medium are provided. An implementation includes: in response to receiving a file moving request, determining a source path and a destination path corresponding to a first moving operation corresponding to the request; determining whether the source path and the destination path corresponding to the first moving operation exist; in response to determining that both the source path and the destination path corresponding to the first moving operation exist, comparing at least one of the source path and the destination path corresponding to the first moving operation with a corresponding path of a source path and a destination path corresponding to the second moving operation; and determining, based on a result of the comparing, whether the first moving operation is to be performed.

Abstract: The present disclosure provides a method for optimizing a tag of a point of interest s(POI). The method includes: obtaining first portrait feature data of each POI in a plurality of POIs and second portrait feature data of each tag in a plurality of marked tags corresponding to the plurality of POIs; mapping the first portrait feature data of each POI and the second portrait feature data of each tag to a metric space to obtain a first feature vector of each POI and a second feature vector of each tag; and optimizing at least one marked tag corresponding to a target POI based on a vector similarity between a first feature vector of the target POI and a second feature vector of at least one tag. The present disclosure provides an apparatus for optimizing a tag of a POI, an electronic device and a computer readable medium.

Abstract: The present application discloses a text processing method and device based on natural language processing and a knowledge graph, and relates to the in-depth field of artificial intelligence technology. A specific implementation is: an electronic device uses a joint learning model to obtain a semantic representation, which is obtained by the joint learning model by combining knowledge graph representation learning and natural language representation learning, it combines a knowledge graph representation learning and a natural language representation learning, compared to using only the knowledge graph representation learning or the natural language representation learning to learn semantic representation of a prediction object, factors considered by the joint learning model are more in quantity and comprehensiveness, so accuracy of semantic representation can be improved, and thus accuracy of text processing can be improved.

Abstract: A mini program batch processing method and apparatus, an electronic device, and a readable storage medium are provided. A method includes: establishing a batch package submission creation event; creating, in response to the batch package submission creation event, a batch package submission task to generate a package submission work queue comprising a plurality of mini program package submission tasks; extracting a first quantity of mini program package submission tasks from the package submission work queue; creating a virtual mini program package for each of the first quantity of mini program package submission tasks; and updating the virtual mini program package to a mini program online package and automatically releasing the mini program online package.

Abstract: The present disclosure provides a method of matching lane line data, which relates to a field of automatic driving, and in particular to a field of locating an autonomous vehicle. The method of matching lane line data includes: obtaining a first data indicating sensed lane lines and a second data indicating map lane lines respectively; constructing a correlation weight matrix for the first data and the second data, based on differences between the sensed lane lines and the map lane lines; and matching the first data with the second data according to the correlation weight matrix, so as to obtain a third data having an optimal match with the first data from the second data. The present disclosure further discloses an apparatus of matching lane line data, a device, a storage medium and a computer program product.

Abstract: The disclosure provides a method for deduplicating entity nodes in a graph database. The method includes: obtaining a set of entity nodes to be deduplicated from the knowledge graph, in which the set includes a plurality of entity nodes; selecting an untraversed entity node from the set as a target entity node; determining a range located by a node identifier corresponding to the target entity node; determining the target entity node that has appeared in traversed entity nodes according to a deduplicating mode corresponding to the range; and deleting the target entity node from the set.

Abstract: A method and an apparatus for generation a high definition map, a device and a computer storage medium, which relate to automatic driving and deep learning technologies in the field of artificial intelligence technologies, are disclosed. An implementation includes: acquiring point cloud data and front-view image data which are collected respectively by a plurality of collecting devices at a plurality of location points to obtain a sequence of point clouds and a sequence of front-view images; performing registration of the front-view images and the point clouds on the sequence of point clouds and the sequence of front-view images; transforming the sequence of front-view images into a top-view image based on the result of the registration and determining coordinate information of each pixel in the top-view image; and identifying map elements of the top-view image to obtain the high definition map.

Abstract: Provided are a method of generating a note, an electronic device, and a storage medium, which relate to a field of artificial intelligence, and in particular to fields of information processing and e-book technology. The method includes: displaying, in a process of displaying an e-book reading page, at least one note combination option for a first note generation instruction, in response to the first note generation instruction being detected; determining a second note generation instruction associated with a target note combination option, in response to an operation of selecting the target note combination option from the at least one note combination option; and combining a current character content associated with the first note generation instruction with a target character content associated with the second note generation instruction, so as to generate a combined note content.

Abstract: A method and apparatus for encoding a geographic location region as well as a method and apparatus for establishing an encoding model, which relate to big data and deep learning technologies in the field of artificial intelligence technologies are disclosed. An implementation includes: determining a to-be-encoded geographic location region; acquiring at least one kind of geographic function information and at least one kind of surface-feature distribution information of the geographic location region; and inputting the acquired geographic function information and the acquired surface-feature distribution information into an encoding model, the encoding model performing embedding on the geographic function information and the surface-feature distribution information, and fusing vector representations obtained by the embedding to obtain an encoding result of the geographic location region.